The risk of drug-induced injury to the developing fetus is a significant and unavoidable problem in the treatment of women with epilepsy who desire to have children. Carbamazepine has held a unique place in the treatment of such women, because it was believed to be nonteratogenic. However, recent clinical studies have shown this view to be mistaken. Several investigators have speculated that carbamazepine itself is not teratogenic, but that these effects are caused by chemically reactive teratogenic metabolites formed from carbamazepine. Many such metabolites are possible. Through in vivo studies in a mouse model of carbamazepine teratogenicity, we seek to identify the enzyme systems that modulate fetal exposure to reactive teratogenic metabolites of carbamazepine. We will consider: 1) cytochrome(s) P-450, 2) epoxide hydrolase, 3) glutathione-S-transferase, and 4) prostaglandin synthase. The experimental design will include: 1) effects of administration of inducers and inhibitors, cosubstrate depletion, and supplementation on fetal development; 2) indicators of fetal exposure to reactive metabolites such as covalent binding of 14C from 14C-carbamazepine to fetal protein; and 3) in vitro studies of carbamazepine metabolism in fetal and maternal tissues including its activation to reactive metabolites. Chromatographic and spectroscopic techniques will be used to identify likely precursors of reactive metabolites for administration in vivo and for in vitro studies of activation. The possibility of a concerted mechanism in which the fetus activates a maternally formed stable metabolite will be addressed through a pharmacokinetic approach. The long-term objective of the proposal is to yield therapeutic strategies to circumvent or minimize the teratogenic effects of carbamazepine in humans.